| Presentation | 2023-09-08 Interview dialogue system based on the evaluation grid method using large-scale language models Daisuke Miyajima, Fan Zhang, Masashi Sugimoto, Kanon Sasaki, Taisei Kitano, Sho Hashimoto, Noriko Nagata, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | The individual interview method using the evaluation grid method is known as a method for structurally understanding users' needs, however, it takes much time and effort to experiment and analyze. In this study, we apply a large-scale language model to the interview framework and construct an interview dialogue system for the evaluation grid method. We implement an interview dialogue for structuring causal relationships (laddering) using the GPT4 model, and automate the process from interview dialogue to evaluation structure analysis by using BERT and the Evaluation Structure Visualization System (ESV) to create evaluation structure diagrams. We compare and discuss the results of the system with those obtained manually. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Large-scale language model / GPT4 / Evaluation grid method / Generative model / Multi-agent |
| Paper # | MVE2023-20 |
| Date of Issue | 2023-08-31 (MVE) |
| Conference Information | |
| Committee | MVE |
|---|---|
| Conference Date | 2023/9/7(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Kiyoshi Kiyokawa(NAIST) |
| Vice Chair | Sumaru Niida(KDDI Research) |
| Secretary | Sumaru Niida(Otsuma Women's Univ.) |
| Assistant | Hidehiko Shishido(Univ. of Tsukuba) / Atsushi Nakazawa(Kyoto Univ.) / Naoya Tojo(KDDI Research) / Naoki Hagiyama(NTT) / Yuji Tatada(Univ. of Tokyo) |
| Paper Information | |
| Registration To | Technical Committee on Media Experience and Virtual Environment |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Interview dialogue system based on the evaluation grid method using large-scale language models |
| Sub Title (in English) | |
| Keyword(1) | Large-scale language model |
| Keyword(2) | GPT4 |
| Keyword(3) | Evaluation grid method |
| Keyword(4) | Generative model |
| Keyword(5) | Multi-agent |
| 1st Author's Name | Daisuke Miyajima |
| 1st Author's Affiliation | Kwansei Gakuin University(Kwansei Gakuin Univ.) |
| 2nd Author's Name | Fan Zhang |
| 2nd Author's Affiliation | Kwansei Gakuin University(Kwansei Gakuin Univ.) |
| 3rd Author's Name | Masashi Sugimoto |
| 3rd Author's Affiliation | Kwansei Gakuin University(Kwansei Gakuin Univ.) |
| 4th Author's Name | Kanon Sasaki |
| 4th Author's Affiliation | Kwansei Gakuin University(Kwansei Gakuin Univ.) |
| 5th Author's Name | Taisei Kitano |
| 5th Author's Affiliation | Kwansei Gakuin University(Kwansei Gakuin Univ.) |
| 6th Author's Name | Sho Hashimoto |
| 6th Author's Affiliation | Seinan Gakuin University/Kwansei Gakuin University(Seinan Gakuin univ./Kwansei Gakuin Univ.) |
| 7th Author's Name | Noriko Nagata |
| 7th Author's Affiliation | Kwansei Gakuin University(Kwansei Gakuin Univ.) |
| Date | 2023-09-08 |
| Paper # | MVE2023-20 |
| Volume (vol) | vol.123 |
| Number (no) | MVE-180 |
| Page | pp.pp.33-38(MVE), |
| #Pages | 6 |
| Date of Issue | 2023-08-31 (MVE) |